[HEADS-UP] systemd for F14 - the next steps

Matthew Miller mattdm at mattdm.org
Thu Jul 22 03:56:03 UTC 2010

On Thu, Jul 22, 2010 at 05:25:19AM +0200, Lennart Poettering wrote:
> Now, after discussing this over 2years with many folks and reading up on
> launchd and SMF and the opinions on the net, we then distilled of the
> requests a set of good features we wanted to implement. Some of those

That is interesting and good for your project. I know you're intelligent and
an great coder, and I have no problem with assuming you did an excellent job
in your research. However, it is -- and I mean this as no insult -- only
tangentially interesting to Fedora.

> matter to you. Others don't. But to other people other features matter
> and the ones you like don't. And even other people have even other
> priorities. 

This is true in a vacuum. In a concrete setting -- a specific distribution
-- it's important to discuss what the priorities of that specific situation
are, and whether a particular piece of software fits those priorities.

And bluntly: I don't think you've done that. The "benefit to Fedora"
section of the feature page says 1) it's a better design (which is to say
nothing about features or priorities or use cases) and 2) it may eventually
speed up booting but is not actually guaranteed to do that yet.

So, while you may have done a good job of fitting your software to the needs
you see in the world, that doesn't automatically mean that there's a good
match for Fedora. (Again, it may be, but you can't just say "We say it's
compelling, so trust us and be compelled.")

> We don't try to make everybody happy with systemd. We will neither offer
> every feature people might have requested, nor are we going to drop
> features just because others don't see the need. Please accept that. It
> doesn't make the project pointless.

I didn't say it was pointless. In fact, I think I'm on record as saying that
it's an interesting project. But while you don't have to "make everybody
happy" in a general sense, in order to be relevant to Fedora, you *do* need
to make a convincing case that the Fedora Project as a whole will be better
off with your project as the default.

And I think it's _very_ fair to say that you had an interesting idea and are
working to get your interesting idea into Fedora quickly for the benefit and
growth of the interesting idea itself, not necessarily for the benefit of
Fedora or Fedora users. In the happy ideal, those two would coincide and
everyone wins, but it's asking a lot of faith from the user community -- and
hubris and arrogance may not be the best strategy.

Matthew Miller <mattdm at mattdm.org>
Senior Systems Architect -- Instructional & Research Computing Services
Harvard School of Engineering & Applied Sciences

More information about the devel mailing list